Automatic detection of Attention Deficit Hyperactivity Disorder (ADHD) based on the functional Magnetic Resonance Imaging (fMRI) through Deep Learning (DL) is becoming a quite useful methodology due to the curse of-dimensionality problem of the data is solved. Also, this method proposes an invasive and robust solution to the variances in data acquisition and class distribution imbalances. In this paper, a transfer learning approach, specifically ResNet-50 type pre-trained 2D-Convolutional Neural Network (CNN) was used to automatically classify ADHD and healthy children. The results demonstrated that ResNet-50 architecture with 10-k cross-validation (CV) achieves an overall classification accuracy of 93.45%. The interpretation of the results was done via the Class Activation Map (CAM) analysis which showed that children with ADHD differed from controls in a wide range of brain areas including frontal, parietal and temporal lobes.
GasconAGamacheDSt-LaurentDStipanicicA. Do we over-diagnose ADHD in North America? A critical review and clinical recommendations. J Clin Psychol. 2022. Advance online publication. doi:10.1002/jclp.23348.
2.
SnyderSMRuginoTAHornigMSteinMA. Integration of an EEG biomarker with a clinician's ADHD evaluation. Brain Behav. 2015;5(4):e00330. https://doi.org/10.1002/brb3.330.
3.
ArnsMConnersCKKraemerHC. A decade of EEG Theta/Beta ratio research in ADHD. J Atten Disord. 2012;17(5):374-383. doi:10.1177/1087054712460087.
4.
SutcubasiBMetinBKurbanMKMetinZEBeserBSonuga-BarkeE. Resting-state network dysconnectivity in ADHD: A system-neuroscience-based meta-analysis. World J Biol Psychiatry. 2020;21(9):662-672. doi:10.1080/15622975.2020.1775889.
5.
SunYZhaoLLanZJiaXXueS. Differentiating boys with ADHD from those with typical development based on whole-brain functional connections using a machine learning approach. Neuropsychiatr Dis Treat. 2020;16:691-702. doi:10.2147/ndt.s239013.
6.
GoodfellowIBengioYCourvilleA. Deep learning. MIT Press; 2016.
7.
ShanmugamaniR. Deep learning for computer vision: Expert techniques to train advanced neural networks using TensorFlow and Keras. Packt Publishing; 2018.
8.
PoggioTAnselmiF. Visual cortex and deep networks: Learning invariant representations. MIT press; 2016.
9.
SantanaANCifreISantanaCNMontoyaP. Using Deep Learning and Resting-State fMRI to Classify Chronic Pain Conditions. Front Neurosci. 2019;13:1313. doi:10.3389/fnins.2019.01313.
10.
MeszlényiRJBuzaKVidnyánszkyZ. Resting State fMRI Functional Connectivity-Based Classification Using a Convolutional Neural Network Architecture. Front Neuroinform. 2017;11:61. doi:10.3389/fninf.2017.00061.
11.
SubahFZDebKDharPKKoshibaT. A deep learning approach to predict autism Spectrum disorder using multisite resting-state fMRI. Appl Sci. 2021;11(8):3636. doi:10.3390/app11083636.
12.
LiaoDLuH. 2018. Classify autism and control based on deep learning and community structure on resting-state fMRI. 2018 Tenth International Conference on Advanced Computational Intelligence (ICACI). doi:10.1109/icaci.2018.8377471.
13.
HonMKhanNM. 2017. Towards Alzheimers disease classification through transfer learning. 2017 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). doi:10.1109/bibm.2017.8217822.
14.
HuJKuangYLiaoBCaoLDongSLiP. A multichannel 2D convolutional neural network model for task-evoked fMRI data classification. Comput Intell Neurosci. 2019;2019:1-9. doi:10.1155/2019/5065214.
15.
DucNTRyuSQureshiMNChoiMLeeKHLeeB. 3D-Deep Learning based automatic diagnosis of Alzheimer’s disease with joint MMSE prediction using resting-state fMRI. Neuroinformatics. 2019;18(1):71-86. doi:10.1007/s12021-019-09419-w.
16.
QureshiMNOhJLeeB. 3D-CNN based discrimination of schizophrenia using resting-state fMRI. Artif Intell Med. 2019;98:10-17. doi:10.1016/j.artmed.2019.06.003.
17.
VuHKimHJungMLeeJ. fMRI volume classification using a 3D convolutional neural network robust to shifted and scaled neuronal activations. NeuroImage. 2020;223:117328. doi:10.1016/j.neuroimage.2020.117328.
18.
KaufmanJBirmaherBBrentD,et al.Schedule for affective disorders and schizophrenia for school-age children-present and lifetime version (K-SADS-PL): Initial reliability and validity data. J Am Acad Child Adolesc Psychiatry. 1997;36(7):980-988. doi:10.1097/00004583-199707000-00021.
19.
DereboyCSenolSSenerSDereboyF. Conners Kisa Form Oğretmen ve Ana Baba Derecelendirme Olçeklerinin Geçerliği [Validation of the Turkish versions of the short-form Conners' teacher and parent rating scales]. Turk Psikiyatri Dergisi = Turkish J Psychiatry. 2007;18(1):48-58.
20.
MetinBTasZCÇebiM,et al.Reward processing deficits during a spatial attention task in patients with ADHD: An fMRI study. J Atten Disord. 2017;22(7):694-702. doi:10.1177/1087054717703188.
FristonKJAshburnerJFrithCDPolineJBHeatherJDFrackowiakRS. Spatial registration and normalization of images. Hum Brain Mapp. 1995;3(3):165-189.
23.
LiXDvornekNCPapademetrisX,et al.2018. 2–Channel convolutional 3D deep neural network (2CC3D) for fMRI analysis: ASD classification and feature learning. 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018). doi:10.1109/isbi.2018.8363798.
24.
AradhyaAMAshfahaniA. 2019. Deep Network Optimization for rs-fMRI Classification. 2019 International Conference on Data Mining Workshops (ICDMW).https://doi.org/10.1109/icdmw.2019.00022.
25.
AbedMTFatemaUNabilSAAlamMARezaMT. 2020. Alzheimers Disease Prediction Using Convolutional Neural Network Models Leveraging Pre-existing Architecture and Transfer Learning. 2020 Joint 9th International Conference on Informatics, Electronics & Vision (ICIEV) and 2020 4th International Conference on Imaging, Vision & Pattern Recognition (icIVPR). doi:10.1109/icievicivpr48672.2020.9306649.
26.
SaengpetchPPipanmemekapornLKamolsantirojS. 2019. Visual Representation Model for fMRI-based Brain Decoding. Proceedings of the 2019 2nd International Conference on Electronics, Communications and Control Engineering - ICECC 2019. doi:10.1145/3324033.3324051.
27.
ZemanAARitchieJBBracciSBeeckHO. Orthogonal Representations of Object Shape and Category in Deep Convolutional Neural Networks and Human Visual Cortex. Sci Rep. 2020;10(1):2453. doi:10.1038/s41598-020-59175-0.
28.
McgovernALagerquistRGagneDJ,et al.Making the black box more transparent: Understanding the physical implications of machine learning. Bull Am Meteorol Soc.2019;100(11):2175-2199. doi:10.1175/bams-d-18-0195.1.
29.
LeeSYTamaBAMoonSJLeeS. Steel surface defect diagnostics using deep convolutional neural network and class activation map. Appl Sci. 2019;9(24):5449. doi:10.3390/app9245449.
30.
JeonYYoshinoKHagiwaraS,et al.Interpretable and lightweight 3-D deep learning model for automated ACL diagnosis. IEEE J Biomed Health Inform. 2021;25(7):2388-2397. doi:10.1109/jbhi.2021.3081355.
31.
TagarisTSdrakaMStafylopatisA. 2019. High-Resolution Class Activation Mapping. 2019 IEEE International Conference on Image Processing (ICIP). doi:10.1109/icip.2019.8803474.
32.
AgrawalH.Comparative analysis of different convolutional neural network algorithm for Image Classification. International Journal for Research in Applied Science and Engineering Technology. 2020;8(9):1110-1120. doi:10.22214/ijraset.2020.31693.
33.
PalrajP.SiddanG.Deep learning algorithm for classification of cerebral palsy from functional magnetic resonance imaging (fmri)International Journal of Advanced Computer Science and Applications. 2021;12(3):1110-1120. doi:10.14569/ijacsa.2021.0120383.
34.
CaiQLiuXGuoZ. 2018. Identifying Architectural Distortion in Mammogram Images via a Se-DenseNet Model and Twice Transfer Learning. 2018 11th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI). doi:10.1109/cisp-bmei.2018.8633197.
35.
NugrohoASuhartantoH. 2020. Hyper-parameter tuning based on random search for DenseNet optimization. 2020 7th International Conference on Information Technology, Computer, and Electrical Engineering (ICITACEE). doi:10.1109/icitacee50144.2020.9239164.
